youtube_dl/extractor/vk.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_str,
  10     compat_urllib_parse,
  11     compat_urllib_request,
  12 )
  13 from ..utils import (
  14     ExtractorError,
  15     orderedSet,
  16     str_to_int,
  17     unescapeHTML,
  18     unified_strdate,
  19 )
  20
  21
  22 class VKIE(InfoExtractor):
  23     IE_NAME = 'vk.com'
  24     _VALID_URL = r'https?://(?:m\.)?vk\.com/(?:video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|(?:.+?\?.*?z=)?video(?P<videoid>[^s].*?)(?:\?|%2F|$))'
  25     _NETRC_MACHINE = 'vk'
  26
  27     _TESTS = [
  28         {
  29             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
  30             'md5': '0deae91935c54e00003c2a00646315f0',
  31             'info_dict': {
  32                 'id': '162222515',
  33                 'ext': 'flv',
  34                 'title': 'ProtivoGunz - Хуёвая песня',
  35                 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
  36                 'duration': 195,
  37                 'upload_date': '20120212',
  38                 'view_count': int,
  39             },
  40         },
  41         {
  42             'url': 'http://vk.com/video205387401_165548505',
  43             'md5': '6c0aeb2e90396ba97035b9cbde548700',
  44             'info_dict': {
  45                 'id': '165548505',
  46                 'ext': 'mp4',
  47                 'uploader': 'Tom Cruise',
  48                 'title': 'No name',
  49                 'duration': 9,
  50                 'upload_date': '20130721',
  51                 'view_count': int,
  52             }
  53         },
  54         {
  55             'note': 'Embedded video',
  56             'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
  57             'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
  58             'info_dict': {
  59                 'id': '162925554',
  60                 'ext': 'mp4',
  61                 'uploader': 'Vladimir Gavrin',
  62                 'title': 'Lin Dan',
  63                 'duration': 101,
  64                 'upload_date': '20120730',
  65                 'view_count': int,
  66             }
  67         },
  68         {
  69             # VIDEO NOW REMOVED
  70             # please update if you find a video whose URL follows the same pattern
  71             'url': 'http://vk.com/video-8871596_164049491',
  72             'md5': 'a590bcaf3d543576c9bd162812387666',
  73             'note': 'Only available for registered users',
  74             'info_dict': {
  75                 'id': '164049491',
  76                 'ext': 'mp4',
  77                 'uploader': 'Триллеры',
  78                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
  79                 'duration': 8352,
  80                 'upload_date': '20121218',
  81                 'view_count': int,
  82             },
  83             'skip': 'Requires vk account credentials',
  84         },
  85         {
  86             'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
  87             'md5': '4d7a5ef8cf114dfa09577e57b2993202',
  88             'info_dict': {
  89                 'id': '168067957',
  90                 'ext': 'mp4',
  91                 'uploader': 'Киномания - лучшее из мира кино',
  92                 'title': ' ',
  93                 'duration': 7291,
  94                 'upload_date': '20140328',
  95             },
  96             'skip': 'Requires vk account credentials',
  97         },
  98         {
  99             'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
 100             'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
 101             'note': 'ivi.ru embed',
 102             'info_dict': {
 103                 'id': '60690',
 104                 'ext': 'mp4',
 105                 'title': 'Книга Илая',
 106                 'duration': 6771,
 107                 'upload_date': '20140626',
 108                 'view_count': int,
 109             },
 110             'skip': 'Only works from Russia',
 111         },
 112         {
 113             # removed video, just testing that we match the pattern
 114             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
 115             'only_matching': True,
 116         },
 117     ]
 118
 119     def _login(self):
 120         (username, password) = self._get_login_info()
 121         if username is None:
 122             return
 123
 124         login_form = {
 125             'act': 'login',
 126             'role': 'al_frame',
 127             'expire': '1',
 128             'email': username.encode('cp1251'),
 129             'pass': password.encode('cp1251'),
 130         }
 131
 132         request = compat_urllib_request.Request('https://login.vk.com/?act=login',
 133                                                 compat_urllib_parse.urlencode(login_form).encode('utf-8'))
 134         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
 135
 136         if re.search(r'onLoginFailed', login_page):
 137             raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
 138
 139     def _real_initialize(self):
 140         self._login()
 141
 142     def _real_extract(self, url):
 143         mobj = re.match(self._VALID_URL, url)
 144         video_id = mobj.group('videoid')
 145
 146         if not video_id:
 147             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 148
 149         info_url = 'http://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
 150         info_page = self._download_webpage(info_url, video_id)
 151
 152         ERRORS = {
 153             r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
 154             'Video %s has been removed from public access due to rightholder complaint.',
 155
 156             r'<!>Please log in or <':
 157             'Video %s is only available for registered users, '
 158             'use --username and --password options to provide account credentials.',
 159
 160             r'<!>Unknown error':
 161             'Video %s does not exist.',
 162
 163             r'<!>Видео временно недоступно':
 164             'Video %s is temporarily unavailable.',
 165         }
 166
 167         for error_re, error_msg in ERRORS.items():
 168             if re.search(error_re, info_page):
 169                 raise ExtractorError(error_msg % video_id, expected=True)
 170
 171         m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page)
 172         if m_yt is not None:
 173             self.to_screen('Youtube video detected')
 174             return self.url_result(m_yt.group(1), 'Youtube')
 175
 176         m_rutube = re.search(
 177             r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page)
 178         if m_rutube is not None:
 179             self.to_screen('rutube video detected')
 180             rutube_url = self._proto_relative_url(
 181                 m_rutube.group(1).replace('\\', ''))
 182             return self.url_result(rutube_url)
 183
 184         m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
 185         if m_opts:
 186             m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
 187             if m_opts_url:
 188                 opts_url = m_opts_url.group(1)
 189                 if opts_url.startswith('//'):
 190                     opts_url = 'http:' + opts_url
 191                 return self.url_result(opts_url)
 192
 193         data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
 194         data = json.loads(data_json)
 195
 196         # Extract upload date
 197         upload_date = None
 198         mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
 199         if mobj is not None:
 200             mobj.group(1) + ' ' + mobj.group(2)
 201             upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
 202
 203         view_count = str_to_int(self._search_regex(
 204             r'"mv_views_count_number"[^>]*>([\d,.]+) views<',
 205             info_page, 'view count', fatal=False))
 206
 207         formats = [{
 208             'format_id': k,
 209             'url': v,
 210             'width': int(k[len('url'):]),
 211         } for k, v in data.items()
 212             if k.startswith('url')]
 213         self._sort_formats(formats)
 214
 215         return {
 216             'id': compat_str(data['vid']),
 217             'formats': formats,
 218             'title': unescapeHTML(data['md_title']),
 219             'thumbnail': data.get('jpg'),
 220             'uploader': data.get('md_author'),
 221             'duration': data.get('duration'),
 222             'upload_date': upload_date,
 223             'view_count': view_count,
 224         }
 225
 226
 227 class VKUserVideosIE(InfoExtractor):
 228     IE_NAME = 'vk.com:user-videos'
 229     IE_DESC = 'vk.com:All of a user\'s videos'
 230     _VALID_URL = r'https?://vk\.com/videos(?P<id>[0-9]+)(?:m\?.*)?'
 231     _TEMPLATE_URL = 'https://vk.com/videos'
 232     _TEST = {
 233         'url': 'http://vk.com/videos205387401',
 234         'info_dict': {
 235             'id': '205387401',
 236         },
 237         'playlist_mincount': 4,
 238     }
 239
 240     def _real_extract(self, url):
 241         page_id = self._match_id(url)
 242         page = self._download_webpage(url, page_id)
 243         video_ids = orderedSet(
 244             m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page))
 245         url_entries = [
 246             self.url_result(
 247                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
 248             for video_id in video_ids]
 249         return self.playlist_result(url_entries, page_id)